Enhancing volunteer engagement to achieve desirable outcomes: what can nonprofit employers do?
Article (Accepted Version)
http://sro.sussex.ac.uk
Alfes, Kerstin, Shantz, Amanda and Bailey, Catherine (2016) Enhancing volunteer engagement to achieve desirable outcomes: what can non-profit employers do? Voluntas: International Journal of Voluntary and Nonprofit Organizations, 27 (2). pp. 595-617. ISSN 0957-8765
This version is available from Sussex Research Online: http://sro.sussex.ac.uk/id/eprint/54259/
This document is made available in accordance with publisher policies and may differ from the published version or from the version of record. If you wish to cite this item you are advised to consult the publisher’s version. Please see the URL above for details on accessing the published version.
Copyright and reuse: Sussex Research Online is a digital repository of the research output of the University.
Copyright and all moral rights to the version of the paper presented here belong to the individual author(s) and/or other copyright owners. To the extent reasonable and practicable, the material made available in SRO has been checked for eligibility before being made available.
Copies of full text items generally can be reproduced, displayed or performed and given to third parties in any format or medium for personal research or study, educational, or not-for-profit purposes without prior permission or charge, provided that the authors, title and full bibliographic details are credited, a hyperlink and/or URL is given for the original metadata page and the content is not changed in any way.
26
Enhancing volunteer engagement to achieve desirable outcomes:
What can nonprofit employers do?
Abstract
Engagement is a positive psychological state that is linked with a range of beneficial
individual and organizational outcomes. However, the factors associated with volunteer
engagement have rarely been examined. Data from 1064 volunteers of a wildlife charity in
the United Kingdom revealed that both task and emotionoriented organizational support
were positively related to volunteer engagement, and volunteer engagement was positively
related to volunteer happiness and perceived social worth and negatively related to intent to
leave the voluntary organization. Consistent with theory, engagement acted as a mediator
between these factors. The implications for future research and the relevance of the findings
for voluntary organizations are discussed.
Keywords
Volunteering; volunteer engagement; retention; happiness; social worth.
27
The academic literature on the management of volunteers has tended to focus on
identifying organizational factors designed to increase volunteer participation, motivation,
and retention (Studer & Schnurbein, 2013; Wilson, 2012). This is not surprising given that
voluntary organizations face the conundrum that it is generally much easier for volunteers to
quit their volunteer employer than it is for salaried workers to quit theirs, thus creating what
has been termed the “important challenge” of retaining a voluntary workforce (Garner &
Garner, 2011: 814). Research has also demonstrated that the implementation of certain
institutional factors yields positive benefits for volunteers themselves (Tang, Choi, &
MorrowHowell, 2010; Tang, MorrowHowell, & Hong, 2009), such as the acquisition and
development of skills (Booth, Park, & Glomb, 2009) and increased health and wellbeing
(e.g., Ayalon, 2008; Pillemer, FullerRowell, Reid, & Wells, 2010). This may not only attract
and retain volunteers, but may also have a positive impact on local communities (United
Nations Volunteers, 2012).
While these studies have advanced our knowledge of some factors which can make a
difference for volunteers and their voluntary organizations, little is known about the causal
mechanism that might explain the relationship between organizational factors and positive
outcomes for volunteers (Jenkinson et al., 2013; Lewig, Xanthopoulou, Bakker, Dollard, &
Metzer, 2007). A notable exception is a study by Lewig et al. (2007) which showed that
burnout and connectedness mediated the relationship between job demands and job resources
with health and determination to continue. In the present study, we extend this earlier
research by proposing and testing a model that examines the effect of task and
emotionoriented organizational support on two other dimensions of volunteers’ wellbeing –
their sense of happiness and perceptions of social worth in addition to their turnover
28
intentions. More importantly, our study suggests an alternative mediator which explains the
relationship between organizational support, volunteer wellbeing and turnover intentions,
namely the extent to which volunteers are engaged with their volunteer work tasks.
Volunteer engagement, as used in the present study, is a relatively new concept in the
volunteering literature and is defined as a unique, positive, motivational construct; volunteers
who are engaged with their volunteer role are fulfilled, invested, and energized by their
volunteer tasks and feel able to express their true selves in the performance of their volunteer
work (Kahn, 1990; Schaufeli & Bakker, 2004; Shantz, Saksida, & Alfes, 2014). Volunteer
engagement therefore has a distinct meaning in that it describes the extent to which
volunteers psychologically, rather than physically, engage with their volunteer work and is
different from the engagement or participation of volunteers in voluntary work in a physical
sense.
We contribute to the literature in at least three ways. First, we focus on two facets of
volunteer wellbeing that have rarely been explored in previous volunteering research, namely
happiness and social worth. By focusing on these outcome variables we provide a broader
picture on the ways in which volunteers benefit from dedicating their time to volunteering
activities, and add to the collection of studies which have demonstrated that volunteering is
beneficial for volunteers’ overall satisfaction and wellbeing (e.g., Jenkinson et al., 2013;
Lewig et al., 2007; Pillemer et al., 2010). Examining these outcomes, along with intent to
stop volunteering, is consistent with Huynh et al.’s (2012) argument that there is a close
association between individual level outcomes for volunteers such as improved wellbeing,
and important organizational outcomes, such as retention.
Second, we develop and test a theoretical model to show that organizational factors –
task and emotionoriented support – are associated with enhanced volunteer wellbeing and
29
retention. Previous research has suggested that organizations can foster volunteering by
offering employees the opportunity to participate in volunteering activities (e.g., Booth et al.,
2009; Caligiuri, Mencin, & Jiang, 2013; Grant, 2012; Jones, 2010; Rodell, 2013). This is
usually done as part of corporate volunteerism programs where employers sponsor release
time and regular compensation to enable interested employees to donate their time to a
specific cause. The present paper takes a more focused perspective by exploring
organizational factors that can be implemented by voluntary organizations to encourage
individuals to volunteer in their free time outside formal work commitments. Specifically, we
examine factors that are within a voluntary organization’s control that can positively enhance
volunteer outcomes in particular (Gagné, 2003; Huynh et al., 2012; Studer & Schnurbein,
2013). In this, we respond to the call for research focusing on organizational factors that are
associated with volunteer outcomes (Boezeman & Ellemers, 2007; CraigLees, Harris, &
Lau, 2008).
Third, we contribute to the volunteering literature by analyzing the mechanism
through which task and emotionoriented support influence the outcomes in our study.
Specifically, we suggest that organizational factors induce a motivational process (Huynh et
al., 2012; Lo Presti, 2013) for volunteers such that they become more engaged when those
factors are present, which results in positive outcomes for their health and reduces their intent
to leave the voluntary organization. We base our argumentation on the Job
DemandsResources (JDR) model, a framework that has been used frequently in the paid
employment sector. This enables us to shed new light on the question of whether theoretical
explanations about the motivations of employees in the paid employment sector are
comparable to those of voluntary workers. Hence the present paper contributes to research on
the similarities between both sectors (Boezeman & Ellemers, 2009).
30
Theory Development and Hypotheses
Theoretical Foundations
“Engagement” entered the lexicon of management research with Kahn’s (1990)
ethnographic study of architects and summer camp workers. Kahn defined engagement as the
harnessing of a person’s full self into their work roles, and emphasized the importance of
employee experiences of meaningfulness, safety and availability in driving engagement.
Kahn’s (1990) theory of personal engagement suggests that engagement is a motivational
concept. Individuals who are engaged allocate resources towards their role, and they intensely
and persistently apply these resources to role performance. Moreover, this theory asserts that
supportive organizational contexts yield high levels of engagement, which in turn, leads to
positive outcomes since individuals work within settings where they feel safe to express their
true self and connect with others. For example, Huynh et al. (2012) showed that job
resources, such as a socially supportive work context led to higher levels of engagement
amongst volunteers and Farmer and Fedor (1999) demonstrated that the extent to which
volunteers believe that they receive support from their voluntary organization influences their
attitudes and behaviors.
A second approach to understanding work engagement was proposed by Schaufeli
and Bakker (2004). They argued that work engagement is a “positive, fulfilling, and
workrelated state of mind that is characterized by vigour, dedication and absorption” (p.
295). This definition of engagement is the centrepiece of the JDR model that posits that job
characteristics fall into two general categories (Bakker & Demerouti, 2007; Demerouti,
Bakker, Nachreiner, & Schaufeli, 2001). Job demands describe aspects of a job that require
31
sustained effort and are related to physiological and/or psychological costs. In contrast, job
resources refer to aspects of a job that may: (1) reduce job demands and their associated
costs, (2) are functional to achieve work goals, and/or (3) stimulate personal growth and
learning (Demerouti et al., 2001). The JDR model specifies two processes through which job
demands and job resources unfold; job demands induce a health impairment process, whereas
job resources evoke a motivational process. In the present study, we focus on the
motivational process which is based on the premise that job resources lead to work
engagement, which in turn, is related to a host of positive individual and organizational
outcomes (Bakker & Bal, 2010; Xanthopoulou, Bakker, Demerouti, & Schaufeli, 2009).
What is common to each model of engagement is that organizations have a pivotal
role to play in generating high levels of engagement, by providing employees with both
economic and socioemotional resources. Moreover, models of engagement agree that the
consequences of engagement are beneficial to both individuals and their employing
organization. For example, Shantz et al. (2013) found a link between engagement and three
aspects of individual performance, and Poulsen et al. (2012) showed an association between
engagement and subjective wellbeing.
Engagement is also an important concept in the context of volunteer work because
volunteers freely give their time to a chosen cause, thereby increasing the capacity to fully
employ and express their true selves into their volunteer activities (Shantz et al., 2014).
Volunteer engagement relates to how volunteers carry out their role and has significant
implications for how organizations operate. Although the volunteering literature has
demonstrated the importance of states and attitudes that are similar to engagement, research
involving paid workers has shown that engagement more fully explains a range of outcome
variables compared to other, more passive states, such as job satisfaction and commitment
32
(Rich, LePine, & Crawford, 2010; Schaufeli, 2013). Moreover, volunteer engagement may be
especially relevant given that research in the paid employment context has shown that work
engagement contributes to both employee wellbeing (e.g., Schaufeli, Taris, & van Rhenen,
2008) and reduced turnover intentions (e.g., Alfes, Shantz, Truss, & Soane, 2013).
Antecedents of Volunteer Engagement
Of the few studies that have explored the antecedents of volunteer engagement, most
have focused on individual differences factors, such as autonomy needs, defined as a desire to
freely choose courses of action (Gagné, 2003), and prosocial motivation (Shantz et al., 2014).
Building on the notion that the volunteer organizational environment is more salient for
attitudes and behaviors than individual differences per se (Hustinx et al., 2010), attention has
turned to the notion that organizational factors may influence engagement. Only two studies
have examined this proposition in relation to volunteers. Huynh et al. (2012) found that social
support, performance feedback, and training were positively associated with volunteer
engagement. Gagné (2003) found a moderate association between environments that provide
volunteers with an opportunity to be autonomous, and engagement.
More generally, studies have demonstrated that supportive organizational factors can
positively influence volunteer’s dedication to volunteer work and their willingness to sustain
volunteering involvement, which in some ways might be considered a proxy for engagement.
For example, Cuskelly, Taylor, Hoye and Darcy (2006) showed that volunteer management
practices such as training were associated with higher levels of volunteer retention. Similarly,
33
research on corporate volunteering programs suggests that employees who are supported by
their employer dedicate more time to volunteering (e.g., Booth et al., 2009).
The organizational context is therefore relevant for volunteer engagement because, as
explained by the JDR model, the organization provides resources that reduce the costs of
demanding job conditions, are functional in achieving goals, and stimulate personal growth
(Bakker & Demerouti, 2007; Demerouti et al., 2001). In the present study, we examine two
organizationlevel resources that research has found to be especially pertinent for volunteers,
namely, task and emotionoriented support.
Taskoriented support includes concrete forms of support that assist volunteers in
overcoming problems experienced during the performance of volunteer work (Boezeman &
Ellemers, 2007). It constitutes a resource because it helps volunteers to manage the costs of
taxing job conditions and helps them to successfully accomplish tasks. For instance,
volunteers who prepare food for homeless people are able to anticipate obstacles in the
kitchen, manage the ebb and flow of guests, and correctly follow health and safety rules, to
the extent that the organization provides them with the necessary task support to carry out the
work.
Emotionoriented support is defined as a form of support that elicits positive feelings
(Boezeman & Ellemers, 2007). It is a job resource because it reduces the psychological costs
of demanding job conditions and facilitates personal growth. Providing emotionoriented
support (e.g., encouragement) to volunteers who serve food to homeless people, for example,
alleviates the stress of cooking for a large number of guests, and helps volunteers to manage
emotions that arise when confronted with poverty. The provision of appreciation stimulates
personal growth, as appreciation leads volunteers to feel efficacious in their role (Bandura,
1982).
34
Research in the volunteering literature resonates with these arguments. For instance,
Boezeman and Ellemers (2007) argued that task and emotionoriented support send cues to
volunteers concerning their status within the organization, and that volunteers derive feelings
of respect and pride as a consequence. Similarly, Lo Presti (2013) found that social and task
support are associated with job satisfaction, commitment, and intent to remain. Unlike the
aforementioned studies, we examine the relationship between task and emotionoriented
support with engagement among volunteers. Moreover, we examine these forms of support
via a different theoretical lens by suggesting that the support provided by the voluntary
organization induces a motivational process in that volunteers are more willing to immerse
themselves in their voluntary work. Doing so is important, as previous research has
demonstrated that engagement is a core underlying mechanism which is better able to explain
how factors in the work environment influence individuals’ attitudes and behaviors,
compared to alternative explanations commonly used in organizational research (e. g.
Christian, Garza, & Slaughter, 2001; Rich et al., 2010; Schaufeli & Bakker, 2004). Active
motivational states such as engagement are likely to be particularly important in the context
of volunteer work because volunteers freely give their time without extrinsic motivators.
Drawing on these notions, we suggest that task and emotionoriented support are
resources that are positively associated with volunteers’ personal investment in their
voluntary work in the form of engagement. Accordingly, we hypothesize:
Hypothesis 1: Taskoriented organizational support is positively related to
engagement.
Hypothesis 2: Emotionoriented organizational support is positively related to
engagement.
35
Outcomes of Volunteer Engagement
A handful of studies have lent preliminary support to the notion that high levels of
volunteer engagement yield similar beneficial outcomes to those observed in the private
sector, such as time spent volunteering (Gagné, 2003; Shantz et al., 2014), satisfaction,
commitment, and intent to remain (Vecina, Chacón, Sueiro, & Barrón, 2012), although one
study, by Huynh et al. (2012), yielded mixed results. Even fewer studies have examined the
potential outcomes of engagement in a psychological sense for volunteers themselves,
although some research has focused on the positive outcomes associated with other attitudinal
variables on the part of volunteers. For example, Boezeman and Ellemers (2007, Study 1)
found an association between commitment to the voluntary employer and volunteers’
intentions to remain. Garner and Garner (2011) found a link between individuals’ satisfaction
with support and intent to remain. More widely, it has been noted that participating in
voluntary work is associated with positive feelings for volunteers (Post, 2005), and Tidwell
(2005) showed that volunteers who identify with their voluntary employer’s vision and values
are more satisfied and committed with their volunteering work. These findings resonate with
studies on corporate sponsored volunteering, which demonstrate that volunteer involvement
leads to beneficial outcomes for the volunteers, the voluntary organization, and the employer
who sponsored the volunteering activities (e.g., Booth et al., 2009; Jones, 2010; Rodell,
2013). For example, Caliguri et al. (2013) showed that volunteering assignments which
included meaningful projects, social support within the voluntary organization, and
opportunities for skill development yielded positive benefits for the employer (i.e. higher
levels of employee engagement) the voluntary organization (i.e. sustainable impact of
volunteering project) and the volunteer (i.e. capability development).
36
Positive outcomes at the individual level are important for volunteer employers and
volunteers alike (Fujiwara, Oroyemi, & McKinnon, 2013; Lewig et al., 2007; United Nations
Volunteers, 2012). In order to address these issues, we focus on three potential outcomes of
engagement that are relevant for voluntary organizations and volunteers: perceived social
worth, happiness, and intent to leave.
Perceived social worth is defined as the “self as valued in interpersonal relationships”
(Grant, 2007: 405). Those who invest their time and effort in volunteering are more likely to
perform well and therefore develop a personal identity as a socially valued individual (Grant,
2007). This is more probable for volunteers who experience engagement with their voluntary
work, given engagement’s association with effort and persistence (Schaufeli, 2013).
Similarly, Kahn’s (1990) theory of personal engagement states that individuals who
experience their work as meaningful and engaging are able to express themselves fully and
are therefore are more likely to feel worthwhile, useful and valuable. Hence volunteers who
are engaged with their work experience a strong sense of social worth.
The theory of personal engagement further suggests that individuals who fully invest
themselves in their role performances, expressing their “preferred self” in a way that
promotes selfexpression and connections to others, experience positive affective states
(Kahn, 1990: 700). The fact that engaged workers are intrinsically motivated and find their
work enjoyable (Schaufeli, 2013) may further foster feelings of happiness and reduced
depressive symptoms (Hakanen & Schaufeli, 2012). Thus, engaged volunteers experience
higher levels of happiness than their less engaged peers.
Finally, volunteer engagement is negatively related to intention to leave because
volunteers who are engaged with their tasks are selfdetermined to accomplish work and
persist in the face of challenges (Meyer & Gagné, 2008). Moreover, volunteers who are
37
engaged with their work are fully absorbed in their work and may experience flow
(Csikszentmihalyi, 1990), that is, they find their volunteer work intrinsically enjoyable and
are more likely to stay. This may be particularly relevant in the notforprofit sector, since
volunteer behavior is less easily mandated, and freedom to quit is far greater than in forprofit
firms (Farmer & Fedor, 2001; Leonard, Onyx, & HaywardBrown, 2004). There is some
empirical support for the link between engagement and retention in the voluntary sector
(Lewig et al., 2007). We therefore additionally propose that engagement is associated with
intent to leave the voluntary organization.
Hypothesis 3: Engagement is positively related to (a) perceived social worth, (b)
happiness and (c) negatively related to intent to leave the voluntary organization.
The Mediating Role of Volunteer Engagement
Our first three hypotheses culminate to position engagement as a mediator of the
relationship between task and emotionoriented support as resources, and the three outcome
variables under investigation. In other words, the reason why resources lead to valued
outcomes is because they ignite in volunteers a sense of engagement with their role. This
hypothesis is consistent with Kahn’s (1990) argument that supportive organizational contexts
yield high levels of engagement which in turn leads to positive individual outcomes.
Moreover, a central tenant of the JDR model is that engagement mediates the relationship
between job resources and positive outcomes. Accordingly, we hypothesize that engagement
mediates the relationship between task and emotionoriented support and the three outcome
variables under investigation:
38
Hypothesis 4: Engagement mediates the relationship between taskoriented
organizational support and (a) perceived social worth, (b) happiness and (c) intent to leave
the voluntary organization.
Hypothesis 5: Engagement mediates the relationship between emotionoriented
organizational support and (a) perceived social worth, (b) happiness and (c) intent to leave
the voluntary organization.
Methods
Sample and procedure
The data used in the present study is part of a program of research that explores
volunteers’ attitudes to their volunteering activities. The participants were individuals
volunteering for a large UK wildlife charity. The survey was distributed to 7,008 individuals
who were recorded on the organization’s volunteer list. Volunteers responded to questions
with regards to individual differences characteristics, the extent to which the organization
supports them in their volunteering role, psychological engagement, and the outcomes of
volunteering. Individuals were sent an email that explained the purpose of the study and its
confidentiality, and included a link to the survey. Two weeks after the initial email, a
reminder email was sent to the volunteers.
The final sample included 1,064 volunteers, constituting a response rate of 15.18 per
cent. In order to test whether the individuals who responded to the survey differed
substantially from those who did not respond, we carried out a test as recommended by
Armstrong and Overton (1977). Specifically, we split our sample in two groups according to
the time when the survey was completed. We then carried out independent ttests across the
study variables. The data revealed that there was no significant difference between early and
39
late respondents on any of the study variables. This lends confidence to the argument that
nonresponse bias did not unduly affect our results and that the 15.18 per cent can be
considered representative of the total sample.
The respondents dedicated their time to performing a variety of volunteering tasks
during the prior 12 months to responding to the survey. The most common volunteering tasks
involved practical conservation work (38,809 hours; e.g., land restoration, managing nature
reserves); residential volunteering (20,444 hours); visitor services for people visiting the
charity’s premises (15,477 hours); and administrative work (13,269 hours). Other
volunteering activities included answering surveys about wildlife behavior (8,472 hours),
fundraising (6,116 hours), member recruitment (4,132 hours), and campaigning (616 hours).
The number of hours that individuals volunteered varied across the sample.
Approximately one quarter (23.2 per cent) of the respondents volunteered up to 20 hours a
year, whereas 10 per cent volunteered more than 340 hours per year. The average number of
hours volunteered per year was 146 (SD=224). The final sample comprised 55.6 per cent
men; the average age was 55.34 years (SD = 13.82) and participants had volunteered for the
organization for an average of 5.61 years (SD = 7.72). The majority of respondents were
married or living in a civil partnership (61 per cent) and from a white background (98 per
cent). 50 per cent of the participants indicated that they were retired, and a further 34 per cent
were employed on either a fulltime or parttime basis. The remaining participants were
unemployed and seeking work (5 per cent), looking after family members (3 per cent),
students (3 per cent), out of work due to illness or disability (2 per cent) or did not indicate
their work status (3 per cent). In terms of highest educational qualifications, 16 per cent had a
higher degree such as a PhD, 39 per cent had a degree, 14 per cent indicated that they had
other higher education qualifications, 9 per cent had completed a preuniversity education, 11
40
per cent finished high school, 8 per cent had job related qualifications and the remaining 3
per cent held other qualifications.
Measures
All items were measured on a 7point Likerttype scale ranging from 1 (“strongly
disagree”) to 7 (“strongly agree”).
Taskoriented organizational support. Consistent with Boezeman and Ellemers
(2007), we used two items to measure task oriented organizational support (GalindoKuhn &
Guzley, 2002). An example item is, “The [Organization] assists me sufficiently in my
volunteering activities” (α = .96)
Emotionoriented organizational support. Consistent with Boezeman and Ellemers
(2007), we used two items to measure emotionoriented organizational support
(GalindoKuhn & Guzley, 2002). An example item is, “The [organization] makes me feel that
it appreciates my efforts” (α = .93).
Volunteer engagement. Volunteer engagement was measured with 9 items, following
the approach by Shantz et al. (2014) who adapted Rich et al.’s (2010) engagement scale to
measure volunteer engagement. The scale measures the three dimensions identified by
Kahn’s (1990) theory of personal engagement: physical engagement (3 items, “I exert a lot of
energy when I volunteer”), emotional engagement (3 items, “I am enthusiastic about my
volunteering activities”), and cognitive engagement (3 items, “When I volunteer, I focus a
great deal of my attention on my activities”). In line with research on engagement, the
subscales were combined to measure overall engagement (α = .92).
Perceived social worth. We used Grant’s (2008) twoitem measure of perceived social
worth. An example item is, “I feel that other people value my contributions” (α = .94).
41
Happiness. We used the eightitem Oxford Happiness Questionnaire (Hills & Argyle,
2002). An example item is, “I am satisfied with my life” (α = .85).
Intent to leave the voluntary organization. We adapted a twoitem measure developed
by Boroff and Lewin (1997). An example item is, “I am seriously considering quitting
volunteering at the [organization]” (α = .75).
Results
Descriptive Statistics
Table I presents the means and standard deviations for each scale, and interscale
correlations, for all study variables.
Insert Table I about here
Preliminary Data Analysis
As all our variables were collected from a single source, we carried out a series of
confirmatory factor analyses (CFA), using Maximum Likelihood estimation in AMOS 22.0
(Arbuckle, 2006), to assess the potential influence of common method variance and to
establish discriminant validity of the scales (Podsakoff, MacKenzie, JeongYeon, &
Podsakoff, 2003). We initially tested a full measurement model, in which the three
engagement facets loaded onto a general engagement factor and all other items loaded on to
their respective factors. All factors were allowed to correlate. We used five fit indices to
establish the goodness of fit of our model. For the X2, values of less than 2.5 indicate a good
model fit and values around 5.0 an acceptable fit (Arbuckle, 2006). For the normed fit index
(NFI) and the comparative fit index (CFI), values greater than .95 represent a good model fit
and values greater than .90 an acceptable fit (Bentler, 1990). For the Root Mean Square Error
of Approximation (RMSEA) and the Standardized Root Mean Square Residual (SRMR),
42
values less than .08 indicate a good model fit (Browne & Cudeck, 1993; Hu & Bentler,
1998).
Insert Table II about here
The sixfactor model showed a good model fit (X2 = 547; df = 136; NFI = .96; CFI =
.97; RMSEA = .054; SRMR = .036). Next, sequential X2 difference tests were carried out.
Specifically, the full measurement model was compared to five alternative nested models as
shown in Table II. This analysis was conducted in order to examine whether the items that
make up the constructs under study should be clustered as predicted. For instance, the full
measurement model represents a model in which the constructs under investigation are
distinct, and the items are clustered according to their theoretical constructs. Model A, on the
other hand, represents a model that does not differentiate between task and emotion oriented
support, in that the items that make up those constructs are combined into a single factor. The
results showed that Model A fitted the data significantly worse than the full measurement
model. This indicates that task and emotionoriented support are best treated as distinct
constructs. Likewise, the results of the other models, in comparison with the full
measurement model, revealed that their model fits were significantly worse compared to the
full measurement model (all at p<.001). Specifically, the results showed that three models
which combined both forms of support and volunteer engagement (Model B), perceived
social worth (Model C) and intention to leave (Model D) fitted the data significantly worse
compared to the full measurement model. This suggests that these variables capture different
constructs and should be treated as distinct.
Finally, we introduced an unmeasured latent method factor to our original
measurement model allowing all items to load on to their theoretical constructs, as well as on
43
to the latent method factor. This is done to evaluate the extent to which common method
variance may influence the results. As expected, the fit of the model including the common
methods factor was significantly better (ΔX2 (1) = 56, p<.001). However, there was only a
marginal improvement in the fit indices between both models. The changes of CFI and NFI
value, comparing both models, were 0.005, and the changes in RMSEA and SRMR values
were 0.004 and 0.001, which does not exceed the suggested rule of thumb of 0.05 (Bagozzi &
Yi, 1990). We performed an additional test for common method variance following the
procedure suggested by Widaman (1985) and applied by Williams, Cote and Buckley (1989).
This approach involves a comparison between a null model, a measurement model, a single
method factor model, and a measurement model with an additional method factor. The results
indicated that the common method factor did improve model fit, however, it only accounted
for a relatively small portion of the variance (16.8 %), which is considerably lower than the
amount of common method variance (25%) observed in Williams et al.’s (1989) study.
Finally, we carried out tests to assess the validity of the scales. To assess evidence for
convergent validity, we computed estimates of construct reliability and average variance
extracted (AVE). Construct reliabilities from the CFA results ranged from .62 to .93 and
therefore either approached or exceeded the recommended threshold of .70 suggested by Hair
et al. (2009). AVE values ranged between .44 and .92, approaching or exceeding the
recommended threshold of .50 (Hair et al., 2009). We found evidence for the discriminant
validity of the study constructs using the method described by Fornell and Larcker (1981), as
each construct’s AVE value exceeded the squared correlation between it and each of the other
study constructs. In summary, the results of the additional tests mitigated concerns that the
associations found in the data were unduly influenced by common method variance.
44
Test of Hypotheses
We employed latent variable structural equation modelling using Maximum
Likelihood estimation in AMOS 22.0 (Arbuckle, 2006) to test our theoretical model. To
examine whether engagement mediated the hypothesized relationships, we followed the steps
outlined by Mathieu and Taylor (2006). The procedure compares three alternative models:
saturated, direct effects, and indirect effects models. For the saturated model, paths were
estimated from each independent variable to engagement, perceived social worth, happiness
and intent to leave, and a direct path from engagement to the three outcome variables. The
saturated model provided a good fit for the data (X2 = 551; df = 139; NFI = .96; CFI = .97;
RMSEA = .05; SRMR = .04).
For the direct effects model, direct paths were estimated from each independent
variable to the outcome variables, whereas no path led to or stemmed from engagement. The
indirect effects model estimated direct paths from each independent variable to the mediator
engagement and a direct path from engagement to the three outcome variables. The direct
effects model and the indirect effects model were both nested within the saturated model,
which enabled us to use X2 difference tests to compare the statistical fit of the three models.
Hence, the X2 difference between the direct effects model and the saturated model, as well as
between the indirect effects model and the saturated model, were tested for significance while
accounting for the change in degrees of freedom between the models. The data are shown in
Table III.
Insert Table III about here
The direct effects model showed a weak model fit (X2 = 1085; df = 145; NFI = .92;
CFI = .93; RMSEA = .08; SRMR = .14), and differed significantly from the saturated model
45
(ΔX2 (6) = 534, p<.001). This indicates that at least one support variable has a significant
direct relationship with engagement, or engagement is significantly related with the outcome
variables; this reinforces the importance of the mediator variable, that is, engagement. The
indirect effects model showed a weak model fit (X2= 1286; df = 145; NFI = .90; CFI = .91;
RMSEA = .09; SRMR = .07) and, again, different significantly from the saturated model
(ΔX2 (6) = 735, p<.001). This lack of fit indicates that one or more of the antecedents has a
direct relationship with the outcome variables.
In the next step, we used the indirect effects model as a base and added direct paths
between the support measures and the outcome variables. We first added individual paths
from taskoriented support to perceived social worth, then to happiness, and then to intent to
leave. Next, we added individual paths from emotionoriented support to perceived social
worth, then to happiness and finally to intent to leave. We kept paths in the model if they
were significant and if adding them resulted in a significant improvement of the overall
model fit, as assessed by X2 difference tests. Apart from the effect of emotionoriented
support on happiness, all other direct effects were significant and led to an improved model
fit, as assessed by X2 difference tests. The fit statistics for the final model are presented in
Table III.
Our results demonstrate that taskoriented (ß=.27) and emotionoriented (ß =.24)
support were positively and significantly related to engagement; thus hypotheses 1 and 2
were supported. Engagement, in turn, was positively and significantly related to perceived
social worth (ß =.18), happiness (ß =.23), and negatively related to intent to leave (ß =.23),
providing evidence in support of hypothesis 3. We examined the significance of indirect
effects using the productofcoefficients approach combined with bootstrapping in AMOS
46
22.0. The indirect effects of taskoriented and emotionoriented support on perceived social
worth, happiness and intent to leave were all significant at the p<.01 level.
In addition, taskoriented support was significantly and positively related to perceived
social worth (ß =.33), happiness (ß =.19), and negatively related to intent to leave (ß =.15).
Moreover, emotionoriented support had a positive and significant relationship with
perceived social worth (ß =.27) and a negative and significant relationship with intent to
leave (ß =.63), but no significant relationship with happiness. This implies that the
relationship between emotionoriented support and happiness was fully mediated by
volunteer engagement and that the relationships between emotionoriented support, perceived
social worth and intent to leave as well as the relationships between taskoriented support,
perceived social worth, happiness and intent to leave were partially mediated by engagement.
The standardized estimates of the final model are represented in Figure 1. Thus we found
evidence to fully support hypothesis 5b and our results lent partial support to hypotheses 4a,
b, and c and hypotheses 5 a and c.
Finally, we reestimated the mediation models by adding a latent method factor,
which loaded on the indicators of all constructs, following the procedure by MacKenzie,
Podsakoff and Fetter (1993). We compared the standardized parameter estimates when
common method variance was, and was not controlled for. The results revealed that, while
the strength of some associations changed slightly, the conclusions drawn from the model did
not change.
Insert Figure 1 about here
47
Discussion: Theoretical and Practical Implications
The present study responds to calls to identify factors associated with volunteer
engagement, and specifically for studies that explore the role that various forms of
organizational support can play in fostering engagement (Huynh et al., 2012; Vecina et al.,
2012). Our study contributes to this literature by examining the associations between
taskoriented and emotionoriented organizational support (Boezeman & Ellemers, 2007) and
volunteer engagement, and three outcomes of engagement that are of interest to both
volunteers and voluntary sector organizations: perceived social worth, happiness, and intent
to leave.
Consistent with our expectations, we found that both task and emotionoriented
support were positively associated with volunteer engagement. These findings build on those
of Boezeman and Ellemers (2007), who found a link between these two forms of support and
feelings of respect among volunteers towards their voluntary organization. We extend their
findings in the application of the JDR model to show that task and emotionoriented support
are job resources, which are associated with high levels of volunteer engagement.
Our study is the first to examine the link between these two forms of support and
engagement, and also one of a small number of studies that have examined the role of
organizational factors implemented by the voluntary organization, as distinct from individual
factors, in enhancing volunteer engagement. We thus contribute to the sparse empirical
literature on the antecedents of engagement in the context of voluntary work.
It is noteworthy that task and emotionoriented support – which we showed were
empirically distinct – had similarly strong effects on engagement such that the provision of
each is equally important in igniting volunteers’ enthusiasm and interest in their volunteer
tasks. However, the indirect effects on the dependent variables in the partially mediated
48
relationships differed; notably, emotionoriented support was more strongly related to
turnover intentions than taskoriented support. This suggests that the general appreciation
volunteers receive from their voluntary organization is more important in retaining them
compared to more specific taskfocused support.
In line with Kahn’s (1990) theory of personal engagement, we also found that
volunteer engagement was positively associated with two facets of employee wellbeing –
happiness and perceived social worth. Understanding volunteering’s role in generating
wellbeing is important not just for volunteers themselves, but also for voluntary
organizations. National and international policy makers are becoming increasingly concerned
with the wellbeing of its citizenship, and they view volunteering as an avenue to achieve this
goal. Since many nonprofit organizations are funded through government agencies, it is in
their interest to show how their work increases volunteer wellbeing.
Unlike previous studies in the volunteer literature, which have shown mixed results
with regards to the relationship between engagement and retention (e.g., Huynh et al., 2012),
our data showed a positive relationship between the two. This finding is consistent with
empirical work in the paid context (Schaufeli & Bakker, 2004). Moreover, it appears
theoretically valid and may in fact be stronger in the case of volunteering. This is because
individuals experience a greater degree of choice over their voluntary work than their paid
employment counterparts, and in many instances, they purposely set aside other activities
such as leisure pursuits in order to undertake volunteering.
We expected to find that engagement fully mediated the relationships between task
and emotionoriented support and the three outcome variables. Although we found evidence
of full mediation in the association between emotionoriented support and happiness, in other
associations, our data showed partial mediation. Specifically, we found that the relationship
49
between emotionoriented support, perceived social worth and intent to leave, as well as
between taskoriented support, perceived social worth, happiness and intent to leave were
partially mediated by engagement. This reinforces the notion that both forms of support exert
different effects on the outcome variables in our study. To an extent, these findings also
reflect those of Huynh et al. (2012). Although they found that connectedness was correlated
with job resources and intent to remain volunteering to approximately the same extent as
engagement was related to those variables, they found that connectedness, and not
engagement, mediated the relationship between the job resources and intention to remain.
The discrepancy in our findings visàvis those of Huynh et al.’s (2012) may be due to
the nature of the samples. In their study, the volunteers worked with AIDS and cancer
patients, where connectedness to other people may be particularly relevant. This is because
the nature of that work likely contains elements of emotional labor, and therefore is
emotionally and mentally depleting. Hence, being engaged with work tasks may be less
central in this context because volunteers may be influenced more by the plight of the
recipients themselves, and focused on responding to them appropriately, in contrast to
focusing solely on their mandated volunteer tasks. In the present study, the volunteers worked
for an environmental nonprofit organization and worked on more physical tasks such as
conservation work; engagement with the task may be more germane here compared to
connectedness given the type of work. We echo Huynh et al.’s (2012) suggestion that future
research should examine the relationship between volunteer engagement and retention across
the type of work that volunteers carry out. Our findings and those of Huynh et al (2012)
imply that there may be additional mediators, aside from engagement, that may be relevant in
a volunteering context. For example, the experience of task and emotionoriented
organizational support might enhance levels of organizational identification which then in
50
turn might mediate the relationship with the outcomes in our study. Further research could
explore such relationships.
A final contribution is that we demonstrate that engagement is a concept that holds in
the nonprofit sector. Indeed, our results complement those found by Lewig et al. (2007) who
used the JDR model to examine the effects of job demands and job resources on volunteers.
They also complement those of Lo Presti (2013) who found that job resources in the form of
social support, task support, appreciation and information were all associated with levels of
commitment, satisfaction and intent to remain. Similarly, in the present study, we found that
the motivational process, as outlined in the JDR model is likewise relevant for nonprofit
organizations. Hence, engagement theory may be useful in unifying the two sectors, and
lessons learned from each may be interchangeable.
From a practical perspective, our research highlights the importance of volunteer
engagement for voluntary sector organizations. Our study suggests that, for those managing
volunteers, the provision of task and emotionoriented forms of support equally enhance
volunteers’ engagement with their voluntary work. Research in the volunteering literature
provides a number of practical suggestions to facilitate these forms of support (e.g., Brudney
& Nezhina, 2005; Hidalgo & Moreno, 2009; Hobson & Heler, 2007). For example,
taskoriented support could be enhanced by providing volunteers with a thorough induction
into the volunteering role, guidelines on how to carry out their role, appropriate resources,
and guidance from other volunteers and paid staff. Doing so may not only benefit the
volunteer organization in ensuring that volunteer hours are maximized, but this form of
support might also benefit volunteers in terms of enhancing their employability, especially
when volunteers are assigned to tasks that can help them develop skills (Booth et al., 2009).
Emotionoriented support could be provided through opportunities to network and meet with
51
other volunteers and paid staff, through feedback and appreciation, access to a mentor, and
awards and recognition.
Finding ways of raising engagement levels amongst volunteers would also be
beneficial. One way to increase volunteer engagement and participation could be to actively
search for opportunities to collaborate with employers from the paid sector. Employees’
involvement in community work outside their organization is positively related to
performance at work (Rodell, 2013). Hence, voluntary organizations could market the
benefits of volunteering to employers and establish joint corporate volunteering programs.
This would not only increase levels of volunteer engagement, but also yield longterm
benefits for the voluntary organization (Caligiuri et al., 2013).
Limitations and Suggestions for Future Research
There are a number of limitations to our study. First, it was based on a selfreport
survey, which raises concerns about common method variance. However, we moderated its
influence by using established scales, and we carried out empirical tests to check for its
influence. These additional analyses lessened concerns that our conclusions were unduly
influenced by common method variance. Nonetheless, studies that draw on data from several
sources (e.g., turnover rates) would further help to alleviate concerns related to common
method variance.
Second, although our hypotheses are based on established theoretical models, our
study is crosssectional which limits our ability to infer causality (Antonakis, Bendahan,
Jacquart, & Lalive, 2010). This is specifically relevant, as our theoretical model involves
mediating relationships. Therefore, the relationships in the present study should be
52
interpreted as correlations. In addition, we encourage future studies to employ longitudinal
research designs to test the mechanism proposed in the present study.
Third, our sample involved volunteers working for a single wildlife charity based in
the UK. We therefore cannot be sure that our findings would hold true in other charitable
sectors or in other national contexts. Further research that explores the role played by
engagement with a wider sample and in different contexts would help show whether the
findings are generalizable.
Finally, our analyses were limited to an examination of two resources as predictors of
engagement. Future studies could explore the association between other forms of resources
and engagement. A preliminary qualitative phase may be useful in order to identify predictors
that are salient to the organizational context. In addition, research could examine personal
resources (e.g., hope, resiliency) as factors that promote volunteer engagement, relying on the
JDR model as a source of theory.
Our study also highlights the similarity of factors affecting work motivations and
outcomes between the voluntary and paid employment sectors. Although our research did not
explore the interface between the two, it remains the case that many volunteers are also paid
employees. The intersection between the work and the volunteering domains is one of great
interest and importance, particularly given research findings that suggest spillover and
compensatory effects between the two spheres of activity (Rodell, 2013). Future research
could extend this body of work by examining the association between work engagement in
the paid versus the voluntary context for individuals, as well as the differences and
similarities in antecedents and outcomes.
53
Conclusion
Voluntary work is of growing economic, political and social importance. The success
of nonprofit organizations relies not just on volunteers being motivated to volunteer in the
first place, but also to sustain their volunteering efforts over time. Our study has shown that
engagement is an important motivational pathway for volunteers, and that the provision of
task and emotionoriented support to volunteers are related to higher engagement levels. We
found that high levels of volunteer engagement are associated with the happiness and
perceived social worth of volunteers, feelings that are likely to enhance the experience of
volunteering, and are of increasing importance to national and international policy shapers.
We also found that engagement was negatively associated with intent to turnover with the
volunteer employer, a desirable outcome for nonprofit organizations. Our research
demonstrates the importance of volunteer engagement, and shows that engagement theory is
relevant not just for paid employees, but for volunteers as well.
TABLE I
Means, Standard Deviations, and Correlations for Scale Variables
M SD 1 2 3 4 5 6 7
1. Gender .44 .50
2. Age 55.34 13.12 .20**
3. Taskoriented organizational support 5.77 1.29 .08** .07*
4. Emotionoriented organizational support 6.06 1.27 .06* .00 .74**
5. Volunteer Engagement 6.01 .79 .04 .11** .34** .31**
6. Happiness 5.69 .72 .15** .02 .27** .20** .28**
54
7. Perceived Social Worth 5.88 1.09 .08* .07* .58** .58** .40** .24**
8. Intent to leave the voluntary organization 5.95 1.23 .04 .01 .63** .73*
* .37**
.22**
.47**
Notes: n = 1064; gender (female=1; male=0); ** p < .01, * p< .05
55
TABLE II
Fit Statistics from Measurement Model Comparison
Models χ²(df) NFI CFI RMSEA SRMR
Full measurement model 547 (136) .958 .968 .054 .036
Model Aa 1328 (141) .899 .909 .089 .047
Model Bb 2204 (145) .833 .842 .116 .087
Model Cc 2586 (145) .804 .812 .127 .063
Model Dd 1489 (145) .887 .897 .094 .048
Model Ee 3886 (145) .705 .712 .157 .170
Model Ff 4573 (148) .653 .660 .169 .144
Model Gg (Harman’s singlefactor test)
6075 (151) .545 .539 .193 .169
Notes: N = 1064, ***p<.001; χ²=chisquare discrepancy, df=degrees of freedom; NFI=Normed Fit Index; CFI=Comparative Fit Index; RMSEA=Root Mean Square Error of Approximation; SRMR= Standardized Root
Mean Square Residual, χ²diff=difference in chisquare, dfdiff =difference in degrees of freedom; in all measurement models, error terms were free to covary between two facets of volunteer engagement to improve fit and help reduce bias in the estimated parameter values (Reddy, 1992). All models are compared to the full measurement model. a=Taskoriented and emotionoriented organizational support combined into one factor b=Taskoriented and emotionoriented organizational support and volunteer engagement combined into one factor c=Taskoriented and emotionoriented organizational support and perceived social worth combined into one factor d=Taskoriented and emotionoriented organizational support and intent to leave combined into one factor e=Happiness, perceived social worth and intent to leave combined into one factor f=Happiness, perceived social worth, intent to leave and volunteer engagement combined into one factor g=All constructs combined into one factor
56
TABLE III
Structural Equation Model Comparisons
Models NFI CFI RMSEA SRMR Comparisons
Saturated model 551 (139) .958 .968 .053 .037
Direct effects model 1085 (145)*** .918 .928 .078 .140 Compared to saturated modelIndirect effects model 1286 (145)*** .902 .912 .086 .065 Compared to saturated modelFinal model 557 (140)** .958 .968 .053 .038 Compared to saturated modelNotes: n = 1064. Error terms were free to covary between two facets of volunteer engagement to improve fit and help reduce bias in the estimated parameter values (Reddy, 1992); *** p < .001, **p < .01
57
FIGURE 1
Standardized Estimates of Final Model
58
References
Alfes, K., Shantz, A. D., Truss, C., and Soane, E. C. (2013). The link between perceived human resource management practices, engagement and employee behaviour: A moderated mediation model. The International Journal of Human Resource Management, 24, 330351.
Antonakis, J., Bendahan, S., Jacquart, P., and Lalive, R. (2010). On making causal claims: A review and recommendations. The Leadership Quarterly, 21, 10861120.
Arbuckle, J. L. (2006). Amos (version 7.0) [computer program], Chicago, SPSS. Armstrong, J. S., and Overton, T. S. (1977). Estimating nonresponse bias in mail surveys.
Journal of Marketing Research, 14, 396402. Ayalon, L. (2008). Volunteering as a predictor of allcause mortality: What aspects of
volunteering really matter? International Psychogeriatrics, 20, 10001013. Bagozzi, R. P., and Yi, Y. (1990). Assessing method variance in multitraitmultimethod
matrices: The case of selfreported affect and perceptions at work. Journal of Applied Psychology, 75, 547560.
Bakker, A. B., and Bal, M. P. (2010). Weekly work engagement and performance: A study among starting teachers. Journal of Occupational and Organizational Psychology, 83, 189206.
Bakker, A. B., and Demerouti, E. (2007). The job demandsresources model: State of the art. Journal of Managerial Psychology, 22, 309328.
Bandura, A. (1982). Selfefficacy mechanism in human agency. American Psychologist, 37, 122147.
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107, 238246.
Boezeman, E. J., and Ellemers, N. (2007). Volunteering for charity: Pride, respect, and the commitment of volunteers. Journal of Applied Psychology, 92, 771785.
Boezeman, E. J., and Ellemers, N. (2009). Intrinsic need satisfaction and the job attitudes of volunteers versus employees working in a charitable volunteer organization. Journal of Occupational & Organizational Psychology, 82, 897914.
Booth, J. E., Park, K. W., and Glomb, T. M. (2009). Employersupported volunteering benefits: Gift exchange among employers, employees, and volunteer organizations. Human Resource Management, 48, 227249.
Boroff, K. E., and Lewin, D. (1997). Loyalty, voice, and intent to exit a union firm, a conceptual and empirical analysis. Industrial & Labor Relations Review, 51, 5063.
Browne, M. W., and Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (eds.), Testing structural equation models, Newbury Park, Sage: pp. 136162.
Brudney, J., and Nezhina, T. (2005). What is old is new again: Achieving effectiveness with volunteer programs in kazakhstan. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 16, 293308.
Caligiuri, P., Mencin, A., and Jiang, K. (2013). Win–win–win: The influence of companysponsored volunteerism programs on employees, ngos, and business units. Personnel Psychology, 66, 825860.
Christian, M. S., Garza, A. S., and Slaughter, J. E. (2001). Work engagement. A quantitative review and test of its relation with task and contextual performance. Personnel Psychology, 64, 89136.
59
CraigLees, M., Harris, J., and Lau, W. (2008). The role of dispositional, organizational and situational variables in volunteering. Journal of Nonprofit & Public Sector Marketing, 19, 124.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience, New York, Harper and Row.
Cuskelly, G., Taylor, T., Hoye, R., and Darcy, S. (2006). Volunteer management practices and volunteer retention: A human resource management approach. Sport Management Review, 9, 141163.
Demerouti, E., Bakker, A. B., Nachreiner, F., and Schaufeli, W. B. (2001). The job demandsresources model of burnout. Journal of Applied Psychology, 86, 499512.
Farmer, S. M., and Fedor, D. B. (1999). Volunteer participation and withdrawal. Nonprofit Management and Leadership, 9, 349368.
Farmer, S. M., and Fedor, D. B. (2001). Changing the focus on volunteering: An investigation of volunteers’ multiple contributions to a charitable organization. Journal of Management, 27, 191211.
Fornell, C., and Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18, 3950.
Fujiwara, D., Oroyemi, P., and McKinnon, E. (2013). Wellbeing and civil society. Estimating the value of volunteering using subjective wellbeing data. Dwp working paper no. 112, Sheffield, Department for Work and Pensions.
Gagné, M. (2003). The role of autonomy support and autonomy orientation in prosocial behavior engagement. Motivation and Emotion, 27, 199223.
GalindoKuhn, R., and Guzley, R. M. (2002). The volunteer satisfaction index. Journal of Social Service Research, 28, 4568.
Garner, J. T., and Garner, L. T. (2011). Volunteering an opinion: Organizational voice and volunteer retention in nonprofit organizations. Nonprofit and Voluntary Sector Quarterly, 40, 813828.
Grant, A. M. (2007). Relational job design and the motivation to make a prosocial difference. Academy of Management Review, 32, 393417.
Grant, A. M. (2008). The significance of task significance: Job performance effects, relational mechanisms, and boundary conditions. Journal of Applied Psychology, 93, 108124.
Grant, A. M. (2012). Giving time, time after time: Work design and sustained employee participation in corporate volunteering. Academy of Management Review, 37, 589615.
Hair, J. F., Black, W. C., Babin, B. J., and Anderson, R. E. (2009). Multivariate data analysis (7 ed.), New Jersey, Pearson Prentice Hall.
Hakanen, J. J., and Schaufeli, W. B. (2012). Do burnout and work engagement predict depressive symptoms and life satisfaction? A threewave sevenyear prospective study. Journal of affective disorders, 141, 415424.
Hidalgo, M. C., and Moreno, P. (2009). Organizational socialization of volunteers: The effect on their intention to remain. Journal of Community Psychology, 37, 594601.
Hills, P., and Argyle, M. (2002). The oxford happiness questionnaire: A compact scale for the measurement of psychological wellbeing. Personality and Individual Differences, 33, 10731082.
Hobson, C. J., and Heler, K. (2007). The importance of initial assignment quality and staff treatment of new volunteers: A field test of the hobsonheler model of nonprofit
60
agency ‘‘volunteerfriendliness’’. International Journal of Volunteer Administration, 14, 47–56.
Hu, L., and Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychological Methods, 3, 424453.
Hustinx, L., Handy, F., Cnaan, R. A., Brudney, J. L., Pessi, A. B., and Yamauchi, N. (2010). Social and cultural origins of motivations to volunteer: A comparison of university students in six countries. International Sociology, 25, 349382.
Huynh, J.Y., Metzer, J., and Winefield, A. (2012). Engaged or connected? A perspective of the motivational pathway of the job demandsresources model in volunteers working for nonprofit organizations. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 23, 870898.
Jenkinson, C., Dickens, A., Jones, K., ThompsonCoon, J., Taylor, R., Rogers, M., et al. (2013). Is volunteering a public health intervention? A systematic review and metaanalysis of the health and survival of volunteers. BMC Public Health, 13, 773.
Jones, D. A. (2010). Does serving the community also serve the company? Using organizational identification and social exchange theories to understand employee responses to a volunteerism programme. Journal of Occupational and Organizational Psychology, 83, 857878.
Kahn, W. A. (1990). Psychological conditions of personal engagement and disengagement at work. Academy of Management Journal, 33, 692724.
Leonard, R., Onyx, J., and HaywardBrown, H. (2004). Volunteer and coordinator perspectives on managing women volunteers. Nonprofit Management and Leadership, 15, 205219.
Lewig, K. A., Xanthopoulou, D., Bakker, A. B., Dollard, M. F., and Metzer, J. C. (2007). Burnout and connectedness among australian volunteers: A test of the job demandsresources model. Journal of Vocational Behavior, 71, 429445.
Lo Presti, A. (2013). The interactive effects of job resources and motivations to volunteer among a sample of italian volunteers. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 24, 969985.
MacKenzie, S. B., Podsakoff, P. M., and Fetter, R. (1993). The impact of organizational citizenship behavior on evaluations of salesperson performance. Journal of Marketing, 57, 7080.
Mathieu, J. E., and Taylor, S. R. (2006). Clarifying conditions and decision points for mediational type inferences in organizational behavior. Journal of Organizational Behavior, 27, 10311056.
Meyer, J. P., and Gagné, M. (2008). Employee engagement from a selfdetermination theory perspective. Industrial and Organizational Psychology, 1, 6062.
Pillemer, K., FullerRowell, T. E., Reid, M. C., and Wells, N. M. (2010). Environmental volunteering and health outcomes over a 20year period. Gerontologist, 50, 594602.
Podsakoff, P. M., MacKenzie, S. B., JeongYeon, L., and Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879903.
Post, S. G. (2005). Altuism, happiness, and health: It's good to be good. International Journal of Behavioral Medicine, 12, 6677.
Poulsen, M. G., Poulsen, A. A., Khan, A., Poulsen, E. E., and Khan, S. R. (2012). Factors associated with subjective wellbeing in cancer workers in queensland. J Med Imaging Radiat Oncol, 56, 347353.
61
Reddy, S. K. (1992). Effects of ignoring correlated measurement error in structural equation models. Educational and psychological measurement, 52, 549–570.
Rich, B. L., LePine, J. A., and Crawford, E. R. (2010). Job engagement. Antecedents and effects on job performance. Academy of Management Journal, 53, 617–635.
Rodell, J. B. (2013). Finding meaning through volunteering: Why do employees volunteer and what does it mean for their jobs? Academy of Management Journal, 56, 12741294.
Schaufeli, W. (2013). What is engagement? In C. Truss, R. Delbridge, K. Alfes, A. Shantz, & E. C. Soane (eds.), Employee engagement in theory and practice, London, Routledge: pp. 1535.
Schaufeli, W. B., and Bakker, A. B. (2004). Job demands, job resources, and their relationship with burnout and engagement: A multisample study. Journal of Organizational Behavior, 25, 293315.
Schaufeli, W. B., Taris, T. W., and van Rhenen, W. (2008). Workaholism, burnout, and work engagement: Three of a kind or three different kinds of employee wellbeing? Applied Psychology: An International Review, 57, 173203.
Shantz, A., Alfes, K., Truss, C., and Soane, E. C. (2013). The role of employee engagement in the relationship between job design and task performance, citizenship and deviant behaviours. International Journal of Human Resource Management, 24, 26082627.
Shantz, A., Saksida, T., and Alfes, K. (2014). Dedicating time to volunteering: Values, engagement, and commitment to beneficiaries. Applied Psychology. An International Review, 63, 671697.
Studer, S., and Schnurbein, G. (2013). Organizational factors affecting volunteers: A literature review on volunteer coordination. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 24, 403440.
Tang, F., Choi, E., and MorrowHowell, N. (2010). Organizational support and volunteering benefits for older adults. Gerontologist, 50, 603612.
Tang, F., MorrowHowell, N., and Hong, S. (2009). Institutional facilitation in sustained volunteering among older adult volunteers. Social Work Research, 33, 172182.
Tidwell, M. V. (2005). A social identity model of prosocial behaviors within nonprofit organizations. Nonprofit Management and Leadership, 15, 449467.
United Nations Volunteers. (2012). Volunteerism and wellbeing, from www.unvolunteers.org/swvr2011
Vecina, M. L., Chacón, F., Sueiro, M., and Barrón, A. (2012). Volunteer engagement: Does engagement predict the degree of satisfaction among new volunteers and the commitment of those who have been active longer? Applied Psychology. An International Review, 61, 130148.
Widaman, K. F. (1985). Hierarchically nested covariance structure models for multitraitmultimethod data. Applied Psychological Measurement, 9, 126.
Williams, L. J., Buckley, M. R., and Cote, J. A. (1989). Lack of method variance in selfreported affect and perceptions at work: Reality or artifact? Journal of Applied Psychology, 74, 462.
Wilson, J. (2012). Volunteerism research: A review essay. Nonprofit and Voluntary Sector Quarterly, 41, 176212.
Xanthopoulou, D., Bakker, A. B., Demerouti, E., and Schaufeli, W. B. (2009). Work engagement and financial returns: A diary study on the role of job and personal resources. Journal of Occupational and Organizational Psychology, 82, 183200.
62